HydroShare: A Platform for Collaborative Data and Model Sharing in Hydrology

Advances in hydrology, as in many domains of science, increasingly requires integration of information from multiple sources, reuse and repurposing of data, and collaboration. The complex, multi-faceted problems faced in hydrology such as predicting floods and droughts in the face of climate and watershed changes cannot be addressed by scientists, either experimentalists or modelers, working individually. Instead, team science and collaboration is required, with data and models open, accessible and transparent to support reproducibility and enhance trust in findings and results. Cyberinfrastructure is needed to help scientists move into this new paradigm of collaborative research. HydroShare is a web based hydrologic information system operated by the Consortium of Universities for the Advancement of Hydrologic Science Inc. (CUAHSI) that is available for use worldwide as a service to the hydrology community. HydroShare includes a repository for users to share and publish data and models in a variety of formats, and to make this information available in a citable, shareable, and discoverable manner. HydroShare also includes tools (web apps) that can act on content in HydroShare, providing users with a gateway to high performance computing and computing in the cloud. This presentation will describe the functionality and architecture of HydroShare, highlighting its use as a virtual research environment for managing individual research contributions within collaborative groups to advance science on complex questions. We will illustrate the use of HydroShare for collecting and making accessible to the community data from the US National Water Model and 2017 Atlantic Hurricanes Harvey, Irma and Maria that had significant impacts on parts of the US and islands in the Caribbean. HydroShare is being used to assemble, document and archive hydrologic data from these events to support research to improve our understanding of, and capability to prepare for and respond to, such extreme events in the future. HydroShare has components that support: (1) resource storage, (2) resource exploration, and (3) web apps for actions on resources. The HydroShare data discovery, sharing and publishing functions as well as HydroShare web apps provide the capability to analyze data and execute models completely in the cloud, overcoming desktop platform limitations. The HydroShare Jupyter Notebook app provides flexible and documentable execution of Python code snippets for analysis and modeling in a way that results can be shared among HydroShare users and groups to support research collaboration. The Jupyter platform is embedded in high performance and data intensive cyberinfrastructure so that code blocks may include preparation and execution of advanced and data intensive models on the host infrastructure. We will discuss how these developments can be used to support collaborative research in hydrology, where being web based is of value as collaborators can all have access to the same functionality regardless of their computer or location. The architecture of HydroShare is built for extensibility with system components loosely coupled and configured to interact through application programming interfaces (APIs). This enhances robustness, as components can be upgraded and advanced relatively independently. Web apps are hosted on separate servers, which may be 3rd party servers set up by different teams. They are registered in HydroShare using a web app resource that configures the connectivity for them to be discovered and launched directly from resource types they are associated with.

Tarboton, D. G., R. Idaszak, J. S. Horsburgh, D. P. Ames, J. L. Goodall, A. Couch, R. Hooper, S. Wang, M. Clark, P. Dash, H. Yi, C. Bandaragoda, A. Castronova, T. Gan, Z. Li, M. Morsy, M. Ramirez, J. Sadler, D. Yin and Y. Liu, (2018), "HydroShare: A Platform for Collaborative Data and Model Sharing in Hydrology," European Geophysical Union General Assembly, Vienna, April 12, Geophysical Research Abstracts 20, EGU2018-9834, https://meetingorganizer.copernicus.org/EGU2018/EGU2018-9834.pdf.

水文领域的进步，如同科学众多领域，日益需要整合来自多个渠道的信息，数据的重复利用和再利用，以及协作。面对水文领域复杂且多面的难题，例如在气候变化和流域变化面前预测洪水和干旱，单靠实验科学家或模型科学家个体的努力是无法解决的。相反，需要团队科学和协作，确保数据和模型公开、易于获取且透明，以支持可重复性和增强对研究结果和结论的信任。为了帮助科学家进入这一新的协作研究范式，需要构建网络基础设施。HydroShare是由大学推进水文科学协会（Consortium of Universities for the Advancement of Hydrologic Science Inc., CUAHSI）运营的一个基于网络的 hydrologic 信息系统，它作为对水文社区的服务，在全球范围内可供使用。HydroShare包括一个存储库，用户可以在其中分享和发布各种格式的数据和模型，并使这些信息以可引用、可分享和可发现的方式可用。HydroShare还包含工具（网络应用程序），可以对HydroShare中的内容进行操作，为用户提供通往高性能计算和云计算的门户。本演示将介绍HydroShare的功能和架构，突出其作为虚拟研究环境在管理协作组内单个研究贡献方面的应用，以推进对复杂问题的科学研究。我们将展示如何使用HydroShare收集和使来自美国国家水文模型和2017年大西洋飓风哈维、艾尔玛和玛利亚的数据可供社区使用，这些飓风对美国部分地区和加勒比海岛屿产生了重大影响。HydroShare正被用于组装、记录和归档这些事件的 hydrologic 数据，以支持研究，提高我们对这些极端事件的理解，以及我们准备和应对这些事件的能力。HydroShare具有支持以下功能的组件：（1）资源存储，（2）资源探索，（3）对资源进行操作的 Web 应用程序。HydroShare的数据发现、共享和发布功能以及 HydroShare 的 Web 应用程序提供的能力，可以在云中完全分析数据和执行模型，克服了桌面平台的限制。HydroShare 的 Jupyter Notebook 应用程序提供了灵活且可记录的 Python 代码片段执行，用于分析和建模，以便结果可以在 HydroShare 用户和组之间共享，以支持研究协作。Jupyter 平台嵌入在高性能和数据处理密集型网络基础设施中，因此代码块可以包括在主机基础设施上准备和执行高级和数据处理密集型模型。我们将讨论如何利用这些发展来支持水文领域的协作研究，其中基于网络的架构具有价值，因为合作者无论其计算机或位置如何，都可以访问相同的功能。HydroShare 的架构旨在可扩展性，其系统组件松散耦合，并通过应用程序编程接口（APIs）进行配置以交互，这增强了其稳健性，因为组件可以相对独立地进行升级和改进。Web 应用程序托管在单独的服务器上，这些服务器可能是由不同团队设置的第三方服务器。它们在 HydroShare 中使用 Web 应用程序资源进行注册，该资源配置了它们的连接，以便可以直接从与之关联的资源类型中查找和启动。